Pediatricians Told Parents to Keep Peanuts Away from Babies Until Age Three. A Trial of 640 Infants Found That Feeding Them Peanuts Early Cut Allergy by 81%.

The Expert Consensus That Backfired

In 2000, the American Academy of Pediatrics told parents of high-risk infants to keep peanuts out of their children's diets until age three. The logic seemed airtight: peanut allergy is severe, infant immune systems are fragile, and exposing babies to a potent allergen felt reckless. Pediatricians relayed the message, preschools banned peanut butter, and an entire generation grew up in peanut-free zones built on the assumption that avoidance equals protection.

Nobody ran a trial. Over the next decade, peanut allergy prevalence in Western countries doubled.

A Clue from the Other Side of the World

The crack in the avoidance theory appeared in 2008, when pediatric allergist Gideon Lack and colleagues published a striking observation: Jewish children raised in the United Kingdom developed peanut allergy at ten times the rate of Israeli children with similar genetic ancestry. The crucial difference was diet. British infants consumed no peanut products in their first year, while Israeli infants ate Bamba, a puffed peanut snack, from roughly seven months of age, averaging 7.1 grams of peanut protein per month. Lack hypothesized that early oral exposure was the protective factor. To test this, he designed the Learning Early About Peanut Allergy (LEAP) trial.

640 Infants, One Clear Answer

The trial enrolled 640 infants aged 4 to 11 months at a single UK site between 2006 and 2009. All participants were at high risk for peanut allergy: they had severe eczema, egg allergy, or both. After a baseline skin-prick test, infants were stratified by peanut sensitivity and randomized to either consume peanut products (at least 6 grams of protein weekly, via Bamba or smooth peanut butter) or avoid them entirely until age 5.

At 60 months, the results were unambiguous. Among infants with no initial sensitivity, peanut allergy developed in 13.7% of the avoidance group versus 1.9% of consumers — an 86% relative risk reduction (P < 0.001). Among those already showing early sensitization, allergy developed in 35.3% of avoiders versus 10.6% of consumers — a 70% reduction (P = 0.004). Adherence reached 92%, and 98.4% of participants completed the trial. A sevenfold difference in allergy prevalence. The protective strategy was the one the guidelines had forbidden.

Why the Gut Overrules the Skin

The LEAP data illuminated a mechanism Lack calls the dual allergen exposure hypothesis: infants with eczema were absorbing peanut proteins from household dust through damaged skin, priming their immune systems to treat peanut as a threat, but oral exposure triggered a different pathway entirely, as the consumption group developed significantly higher peanut-specific IgG4 antibodies, a tolerance marker that mirrors successful immunotherapy, while the avoidance group accumulated peanut-specific IgE, the antibody class behind allergic reactions.

The gut taught tolerance. The skin was teaching war.

Protection That Survived Neglect

Two follow-up studies tested whether the benefit was durable. In the LEAP-On study (2016), 550 participants stopped eating peanuts for a full year after the trial ended, yet at age 6, allergy remained at 4.8% in the original consumption group versus 18.6% in avoiders, a 74% relative reduction that held despite twelve months of zero peanut intake.

Then came the definitive test. LEAP-Trio, published in NEJM Evidence in 2024, followed 508 of the original participants to age 12 or older. During those intervening years, children ate peanuts however they wished, and consumption varied widely. At adolescence, allergy prevalence held at 4.4% in the original consumption group and 15.4% in the avoidance group, a 71% reduction sustained across nearly a decade of unrestricted, inconsistent eating, proving the protection was genuinely durable. Five years of early peanut consumption had created immunological tolerance that lasted.

The Cost of Being Wrong

Roughly 10–15% of the 3.6 million annual US births qualify as high-risk for peanut allergy because they present with severe eczema or egg allergy, yielding 360,000 to 540,000 high-risk infants each year. Applying the LEAP trial's 11.8-percentage-point absolute risk reduction to this population produces approximately 42,000 to 64,000 preventable cases annually. Over the 15 years between the AAP's avoidance guidelines and their reversal, that compounds to between 630,000 and 960,000 cases of peanut allergy that might never have occurred. Nearly a million children. All because the original guideline was issued on expert opinion instead of experimental evidence.

The Strongest Counterargument

The trial excluded 9.1% of screened infants (76 of 834) who already showed strong peanut sensitization — wheals larger than 4 mm on the skin-prick test. These children likely already had peanut allergy, and feeding them peanut products could have triggered anaphylaxis, which means the simple public health message "feed your baby peanuts early," while accurate for most infants, is potentially dangerous for the most sensitized, and the practical challenge of identifying those children at scale without delaying introduction for everyone else remains nontrivial. The AAP's 2017 Addendum Guidelines addressed this with a three-tier system requiring allergist evaluation for the highest-risk infants, but access to allergists varies enormously, and the screening step can itself delay the early introduction the evidence supports.

The EAT trial (n=1,303), which tested early allergen introduction in a general-population cohort, found only a non-significant 20% reduction in food allergy in the intention-to-treat analysis, though significant reductions appeared in the per-protocol analysis. The LEAP trial's 81% figure is specific to high-risk infants; extrapolating it to all babies overstates the current evidence.

What We Didn't Prove

LEAP was a single-site, open-label trial in the UK with limited racial and ethnic diversity, and though the primary outcome was assessed via double-blind oral food challenge (partially mitigating the open-label design), families knew their assignment throughout the five-year intervention period, which could have influenced compliance or reporting. The minimum effective dose was not determined; the consumption group ate at least 6 grams of peanut protein weekly, but whether less would suffice is unknown. The trial says nothing about treating established peanut allergy. And while LEAP-Trio documented protection through age 12, whether tolerance is truly lifelong remains an open question that ongoing follow-up may eventually answer.

What You Can Do

Introduce peanut-containing foods (smooth peanut butter thinned with breast milk or formula, or dissolvable peanut puffs) between 4 and 6 months of age, after your baby has tolerated other solid foods. If your infant has severe eczema or a confirmed egg allergy, consult a pediatrician or allergist first; these children benefit from skin-prick testing before introduction. Never give whole peanuts or chunky peanut butter to children under 5 due to choking risk. Ask your pediatrician about the NIH's three-tier framework from the 2017 Addendum Guidelines to determine whether your infant should be evaluated before or after introduction. Start early: every month of delay narrows the window during which the immune system is most receptive to building tolerance rather than hostility.